matkeane

Hi Benjamin, On a recent project, with both Blackmagic and Datapath live inputs running simultaneously, I added a cue for each of the live inputs to the Main Timeline, on stage within the relevant display, but behind the background image so that they were hidden from view. I extended these two cues to the entire length of the show, so that they were always present. The live inputs for display were added to tasks, and cued up when required. Apart from the live video of the presenter animating into place, there wasn't much other animation on-screen, but I didn't notice any stutter when they came on. Your mileage may vary if you have a lot of other animation going on, I guess. It was the Dataton distributor in Paris who showed me the trick with the hidden cues to keep the inputs active. Without them, I saw between 1-2 seconds of delay before the live video showed, so I always do this now.

Hi, In my experience, there is often a delay until the live video inputs become visible. So your smooth fade is probably working, but because the video isn't being displayed until the fade is almost done, it appears as if it were a cut transition. To avoid this, I try to keep a copy of the live input active by hiding it on a display somewhere behind a background, or by keeping the video task active, but fading the opacity to zero. This seems to prevent the live video inputs going into 'standby' and when I launch a task with a live video cue, it is visible immediately. I don't keep know whether having live inputs running constantly has a performance impact, but it seems to avoid the wait before the live input signal is visible. This probably depends on the capture cards - Datapath cards seem to cue up more swiftly than the Blackmagic ones, for example.

It sounds like he wants to pan and zoom the image, without changing the image bounds - like an animated picture within a frame - in which case a second cue used as a matte would do the job.

Mike, I haven't done any really crazy sizes, as I prefer to keep within multiples of 16px, but I have exported non-standard video sizes - e.g. 704x768, 656x656, 352x128 - in AME. Some 'profile' and 'Level' options restrict exports to standard video sizes and framerates (e.g. 'Simple' profile won't exceed 720x576px), but 'Main' profile with 'High' level allows you more freedom, although with no guarantees that the file will play back on all hardware or software.

Hi Brian, Like you, I know more production companies using the Adobe suite than using Episode, so thanks for taking the time to do those tests with AME and publish your results. By the way, what did you use to check the frame structure of the files after encoding?

Hi, One way to do this is to make a looping composition and place it, with Free Run and Looping selected, in an auxiliary timeline. Now you can add an opacity fade in/out to the media cue in the auxiliary timeline, and add a Pause cue with the opacity at 100%. When you start the auxiliary timeline, your loop will fade on and then pause, while looping, at 100% opacity. When you want to fade out, hit play on the auxiliary timeline.

Re: Brian's idea about rapidly building 'arrays' of displays. I guess one way around this is to duplicate and move the displays one by one - e.g. if you know you have a series of 1980 displays with 300px overlap, then you can duplicate the first display, move the copy 1680px on the x-axis and so on, but it's fiddly if anything changes. Recently, when a client was thinking about a video wall with 20 LCD monitors in a grid, I built a little Processing application that would do the layout maths for me - a bit like you described: choose a resolution, then the number of displays, set the overlap/spacing and generate the array of displays and their coordinates. The next step was to be an export to Watchout, but I never got that far! So yes, I could see something like this being a useful addition to Watchout - the question is how to integrate it into the software without the UI getting too messy. Perhaps something like the 'distribute' command in Adobe apps - position the first and last displays accurately, then automatically distribute the copies, equally spaced between the two? Or the 'repeat' command - choose the number of copies and the x & y offset for each copy (Added bonus if WO can calculate the x-offset from a given overlap)? On a related note, yesterday I spent a while making changes to a project, which involved creating compositions to fit a certain number of displays (rather than the default size which fits all displays). Despite the fact that I'm working on a powerful multi-core computer which could do the calculations in a fraction of a second, this involved a lot of writing numbers of bits of paper, adding things up with calculator, and then typing them back into Watchout! Anything which results in the computer doing more of the dull maths, and me spending more time doing more interesting stuff would be good by me!

Hi Mike, I agree that a separate type of control cue would be the most useful. With media cues, we often use an 'in time' setting and perhaps also fade out before the end, so it would be useful to be able to add a new control cue at an arbitrary point in the timeline and then extend its duration to the length desired. If we could rename the Cue and see the name show up in the Status window alongside the timecode, and maybe have the option to count up or down, I think that would cover most use cases - and be extremely useful. Thanks

I had the same experience as Brian on a recent show. Running WO 5.1 I attempted to name the proxy folder and files using the new, simpler method mentioned in the v5 docs and got the same error. I checked upper/lower case, made sure there were no extra spaces, and got the same error each time. When I gave up and reverted to the v4 naming convention, the proxy import worked first time. This isn't really a reliable bug report, as this happened about 10 minutes before the show when the client arrived with a new media file, but it seems to me that something about the v5 naming method is easier to get wrong.

Copy-pasting the media into a new composition works, but you still have to set up the size and position for each new composition, which quickly gets boring if you're faced with creating over 20 duplicates and don't want to use the default values. Not sure if this is the best way to make a feature request, but a 'duplicate' command would often be very handy.

Hi, One solution might be to set the Z Anchor Position of your image to the desired circle radius in order to push it away from the rotation point. Then if you rotate the image around the Y axis, for example, if moves in a circle - although it won't face the camera. If you need to apply a rotation and have the image face the camera, I guess you could create a composition in which the image makes the opposite rotation, and then move the comp in a circle.

I think this is the sort of situation where the Display settings come into play. When you double-click a display to bring up the settings, you will see an option - checked by default - for 'Width and Height same as display resolution'. Unchecking this allows your display area in Watchout to be different from the resolution of the player & projector. For example, on a recent project with 5 1920x1080 displays, one of the players was unable to drive the projector at the correct resolution. Rather than try to change the show, I was able to set the 'Display resolution' to 1280x720 but tell Watchout to maintain the 1920x1080 area for the display on the stage, and so images that spanned all 5 displays were aligned as intended. I realize that you're trying to achieve the opposite - same resolution, different stage size - to me, but I think the same settings will help you: Your Display Resolutions will be the same, but their stage size will differ.